Problem: A funccal is garbage collected while it can still be used.
Solution: Set copyID in all referenced functions. Do not list lambda
functions with ":function".
bc7ce675b2
Problem: Leaking memory when redefining a function.
Solution: Don't increment the function reference count when it's found by
name. Don't remove the wrong function from the hashtab. More
reference counting fixes.
8dd3a43d75
Problem: Using function() with a name will find another function when it is
redefined.
Solution: Add funcref(). Refer to lambda using a partial. Fix several
reference counting issues.
437bafe4c8
Problem: User defined functions can't be a closure.
Solution: Add the "closure" argument. Allow using :unlet on a bound
variable. (Yasuhiro Matsumoto, Ken Takata)
10ce39a0d5
Problem: Using submatch() in a lambda passed to substitute() is verbose.
Solution: Use a static list and pass it as an optional argument to the
function. Fix memory leak.
df48fb456f
Problem: filter() and map() either require a string or defining a function.
Solution: Support lambda, a short way to define a function that evaluates an
expression. (Yasuhiro Matsumoto, Ken Takata)
069c1e7fa9
Problem: Cannot detect a crash in tests when caused by garbagecollect().
Solution: Add garbagecollect_for_testing(). Do not free a job if is still
useful.
ebf7dfa6f1
1. When calling writefile(list, fname, []) do not show error message twice.
2. Do not allow file name to be overwritten for writefile([1], 2).
3. Do not show “Can’t open file with an empty name” error after error like
“using Float as a String” when type of the second argument is not correct.
4. Do not give multiple error messages and still continue for code like
`writefile(["test", [], [], [], "tset"])`.
Note that to fix 4. ideally I need tv_check_str_or_nr which is currently present
in two PRs: #6114 and #5119. I would want to avoid copying this function into
a yet another PR.
Ref vim/vim#1476.
The exists('g:loaded_foo') convention is rather common, and may be
relied upon in some cases. It's also very unlikely that a user or plugin
has any reason to set g:loaded_foo to zero, so the principle of least
surprise can be brushed aside here.
https://github.com/neovim/neovim/issues/6107#issuecomment-279532143
According to the documentation fsync() may fail with EROFS or EINVAL if “file
descriptor is bound to a special file which does not support synchronization”
(e.g. /dev/stderr). This condition is completely valid in this case since main
point of `file_fsync()` is dumping buffered input.
- Do not exclude any directories from `find` search, remove dumps before tests
instead.
- Install `apport` on travis so that linux tests should produce core dumps
(based on information from travis-ci/travis-ci#3754, not sure whether it still
applies).
- Check cores in lua so that one has an idea which test is failing exactly. Do
this only 10% of time on linux because traversing the file system is slow.
Unit tests are still not touched, though it is what `app` argument in
`check_cores` is for.
TODO? consider using `find`, it may be faster. Consider retiring `os.execute`,
dealing with escaping is bad.
We must invoke src/nvim/testdir/Makefile directly.
Explained in 3d1084f264:
> Running tests from the top-level Makefile will use the third-party
> dependencies from .deps instead of the ones from the Travis cache.
If we could run `oldtest` with CMake, we would not need to do this. Need
USES_TERMINAL feature (CMake 3.2+) for that.
